You can use generated results for commercial purposes. You own all rights to the content you create. The only restriction: make sure your prompt does not include copyrighted third-party material. You are responsible for respecting the rights to any input data.

This is a speech synthesis model from Google: it converts text to audio. According to our data, it accepts up to 65,536 tokens as input, so you can send large amounts of material at once — an article, a podcast script, assistant lines, voiceovers for videos, or training courses.

We do not specify exact file parameters — frequency, bitrate, container — in the model card; please refer to the API response. What is known from the data: the model provides up to 32,000 output tokens per request, and the result comes as an audio track suitable for editing.

There is no indication of reasoning before answering in our data, which is expected for a voiceover task: the model does not solve logical problems but converts ready-made text into speech. If you need a chain of thought, use a text model in BotHub and send its result for synthesis.

There is no mention of function calling or structured JSON output for this model in our data. Integration is not complicated by this: requests go through the unified OpenAI-compatible BotHub API, and for a strict response format in your scenario, let a text model connected with the same key handle it.

Inputting anything other than text is not noted in our data: the main scenario is text input and audio output. If your project requires transcription or image analysis, connect a model with suitable capabilities in the same BotHub window.

We do not specify the language of prompts and voiceovers in the model data, so we cannot promise specific quality — please test it with your own text fragment. In BotHub, this is available without a VPN or foreign card, payment is in rubles upon usage, and unused tokens do not expire.